Visual Tools
Calculators
Tables
Mathematical Keyboard
Converters
Other Tools

How to Find Eigenvalues

The eigenvalues of a square matrix are the numbers λ\lambda for which Av=λvA\mathbf{v} = \lambda\mathbf{v} has a nonzero solution. Set the characteristic polynomial equal to zero and solve; the roots, counted with multiplicity, are the eigenvalues.

Recognize the task

A square matrix and the words eigenvalues, spectrum, or "for which λ\lambda is A−λIA - \lambda I singular?": (5412)\begin{pmatrix} 5 & 4 \\ 1 & 2 \end{pmatrix}, a triangular matrix, a 3×33 \times 3 with a sparse column. Test: you need numbers λ\lambda, not vectors; the vectors come next, in finding eigenvectors.
Eigenvalues are defined in the Definition section of the eigenvalue lesson; the equation to solve is the characteristic equation.

Ways to do it

Characteristic equationany size
det⁡(A−λI)=0\det(A - \lambda I) = 0
Build the polynomial as in finding the characteristic polynomial, then factor it.
2×2 quadratic formula2×2
λ=tr⁡(A)±tr⁡(A)2−4det⁡(A)2\lambda = \frac{\operatorname{tr}(A) \pm \sqrt{\operatorname{tr}(A)^2 - 4\det(A)}}{2}
The sign of the discriminant classifies the eigenvalues: two real, one repeated, or a complex pair.
Triangular or diagonalinspection
λi=aii\lambda_i = a_{ii}
The eigenvalues are the diagonal entries; no algebra is needed.

Steps

StepOn A=(5412)A = \begin{pmatrix} 5 & 4 \\ 1 & 2 \end{pmatrix}
1Look for a shortcut. If the matrix is triangular or diagonal, read the diagonal and stop. Otherwise continue.not triangular
2Form the characteristic polynomial. Use λ2−tr⁡(A)λ+det⁡(A)\lambda^2 - \operatorname{tr}(A)\lambda + \det(A) for 2×22 \times 2; expand det⁡(A−λI)\det(A - \lambda I) otherwise.tr⁡(A)=7\operatorname{tr}(A) = 7, det⁡(A)=6\det(A) = 6: p(λ)=λ2−7λ+6p(\lambda) = \lambda^2 - 7\lambda + 6
3Set it to zero and solve. Factor, use the quadratic formula, or for a cubic test the divisors of the constant term with the rational root test.(λ−1)(λ−6)=0(\lambda - 1)(\lambda - 6) = 0
4List the roots with their multiplicities. A root repeated kk times has algebraic multiplicity kk. Complex roots count too.λ1=6\lambda_1 = 6, λ2=1\lambda_2 = 1, each once
5Check with the trace and determinant. The eigenvalues add up to the trace and multiply to the determinant.6+1=76 + 1 = 7 ✓, 6⋅1=66 \cdot 1 = 6 ✓

Worked examples

Example 1 · a triangular matrix

inspection
A=(308−1)A = \begin{pmatrix} 3 & 0 \\ 8 & -1 \end{pmatrix} is lower triangular
zero above the diagonal
λ=3, −1\lambda = 3, \ -1
read the diagonal

Example 2 · quadratic formula

2×2
A=(1331)A = \begin{pmatrix} 1 & 3 \\ 3 & 1 \end{pmatrix}: tr⁡(A)=2\operatorname{tr}(A) = 2, det⁡(A)=−8\det(A) = -8
λ=2±4+322=2±62\lambda = \frac{2 \pm \sqrt{4 + 32}}{2} = \frac{2 \pm 6}{2}
discriminant 36>036 > 0: two real roots
λ=4, −2\lambda = 4, \ -2; check 4+(−2)=24 + (-2) = 2, 4⋅(−2)=−84 \cdot (-2) = -8
trace and determinant agree

Example 3 · expand along a sparse column

3×3
A=(401−210−201)A = \begin{pmatrix} 4 & 0 & 1 \\ -2 & 1 & 0 \\ -2 & 0 & 1 \end{pmatrix}
column 2 has two zeros
det⁡(A−λI)=(1−λ)[(4−λ)(1−λ)+2]\det(A - \lambda I) = (1 - \lambda)\left[(4 - \lambda)(1 - \lambda) + 2\right]
expand along column 2
=(1−λ)(λ2−5λ+6)=(1−λ)(λ−2)(λ−3)= (1 - \lambda)(\lambda^2 - 5\lambda + 6) = (1 - \lambda)(\lambda - 2)(\lambda - 3)
λ=1,2,3\lambda = 1, 2, 3; sum 6=tr⁡(A)6 = \operatorname{tr}(A), product 6=det⁡(A)6 = \det(A)
both checks pass

See it happen

Most vectors change direction when AA acts on them. An eigenvalue is the stretch factor along one of the rare directions that do not turn: AvA\mathbf{v} lands on the same line as v\mathbf{v}, scaled by λ\lambda. A negative λ\lambda flips the vector, and λ=0\lambda = 0 crushes that direction to the origin. Drag a vector around in the eigenvectors 2D visualizer and watch it line up with its image only along the eigen-directions.

Common mistakes

Row reducing before finding eigenvalues

✗ reduce (5412)\begin{pmatrix} 5 & 4 \\ 1 & 2 \end{pmatrix} to (5406/5)\begin{pmatrix} 5 & 4 \\ 0 & 6/5 \end{pmatrix} and read 55 and 6/56/5
✓ row operations change the eigenvalues; solve det⁡(A−λI)=0\det(A - \lambda I) = 0: λ=6,1\lambda = 6, 1

Dropping a repeated root

✗ p(λ)=−(λ−1)2(λ−4)p(\lambda) = -(\lambda - 1)^2(\lambda - 4) gives the eigenvalues 1,41, 4 and nothing more
✓ list 1,1,41, 1, 4: λ=1\lambda = 1 has multiplicity 22, and 1+1+4=61 + 1 + 4 = 6 is the trace

Discarding complex roots

✗ λ2+1=0\lambda^2 + 1 = 0, so there are no eigenvalues
✓ λ=±i\lambda = \pm i, a conjugate pair; see complex eigenvalues

Reading the diagonal of a non-triangular matrix

✗ the eigenvalues of (1331)\begin{pmatrix} 1 & 3 \\ 3 & 1 \end{pmatrix} are 1,11, 1
✓ only triangular matrices show their eigenvalues on the diagonal; here λ=4,−2\lambda = 4, -2